Analysis of the Effects of Fixed Costs on Learning Curve Calculations

摘要

The goal of this research was to analyze the effect of fixed costs on learningcurve calculations. The research focused on two general research areas. The first area addressed the identification of current cost analysis practice for handling fixed costs in learning curve calculations. After identification of current practice, the unit learning curve model's (AX(b)) ability to estimate total production run costs for a new production run when the slope was derived from both total cost and variable cost historical lot data was examined. Historical lot cost data was simulated for both the total cost and variable cost cases under three slopes, three fixed cost percentages, and three lot sizing profiles. The second area addressed the predictive ability, measured by mean absolute deviation (MAD), of the AX(b) model versus the SAF/FMC model (F/Q + AX(b) when fit to total cost lot data. Comparisons between both models under different conditions were addressed through ANOVA. Three main findings: (1) there are a variety of practices when handling fixed costs in learning curve calculations, (2) the AX(b) model by itself was inadequate for production run total cost estimating, and (3) the SAF/FMC model was superior to the AX(b) model under all conditions. Fixed costs, Variable costs, Learning curve, Recurring costs, Nonrecurring costs, Unit theory, Linear regression, Overhead, Cost models.